Future request functionalities

[Ole Marius Svendsen]

After using the VI-Suite for daylight studies i would like to bring up some request which i do think would improve your tool. I hope some of these can be implemented one day.

1. Allow "/" 
As we have discussed in the first post i made, you assisted in the debugging that VI-Suite do no allow the "/", and thus no material can be detected. As a commen worktool now adays is IFC files which many consultans receive, or even with the BlenderBim tool, beeing in a native IFC Language requires "/". I would hope this bug may be easy to resolve as i can see the benefit of being allowed to author the IFC file, whilst conducting simulations studies.

2. Blender file and IFC file.
In this section, i will just let you know i want to address question on the OSA forum if they know a easy way to integrate blend files and IFC files. As it stands now with the BlenderBim tool, they save their files in IFC format as saving it to Blender format is invalidating the IFC Format. I would hope they have a solutions so that the VI-Suite materials can be saved into the IFC format so any parametric auditing of their wall/window/floor/slap/roof resources can be done. As it stands now when loading the IFC file, and saving to blender i can no longer use parametric designs for their IFC files as i have to be in blender file for your studies. 

3. Results in tables and excel export
Whilst running a set of simulations, i notice that the VI Metrics has a output slot. I have yet to find a way to export all my results into a table. For the large scale project we have, it can at times be 100 rooms simulated at the same time. Having then an option for exporting the results to a excel, or open office format would be highly requested from my side. The way i understand VI Metric, is that i need to change zone by zone to see its results. I also think there should come up a questions for BREEAM results if the users have excluded 0,5 m from the facade as this is the requirement from the BREEAM manual. 

4. 3D view for reflection factors

When doing a peer review to ensure the quality for the settings has been done accurately, this has always been faster when seeing it in a visual context. I therefore suggest a view in 3D view where the reflection factors is shown.

5. Climate based daylight with simplified blinds
Yesterday i was working on a project where we wanted to use sDA as a mean of comparing two set of alternatives. The way we wanted is to showcase when having an existing facade you would need blinds to avoid both overheating and glare. Thus we used sDA with blinds that activates when the window has a certain W/m2 sun insolation on it, as shown in an example week for the project underneath. Would it be possible to add a simplefied version of this within your tool?



6. Annual Glare studies.
As mentioned in the previous tread, i hope one day we can see a Annual Glare studie, much like Climate Studio has for Rhino. I hope this would one day find its way to VI-Suite.

7. vertical sky component

We often use VSC for early studies, or to find average obstruction angles for windows in our national building codex. For this we use DE LUMINAE plugin for Sketchup. I hope VSC one day may find its way into your tool. If it do, then again back to the comment of having tables would be apricated. 

[VI-Suite]

Regrading 7, VSC is basically the percentage of the sky that can be seen so can't you just use the Sky View Factor node?

Sky View Factors and other simulation results can be written out to CSV with the VI CSV node. This does not yet apply to average DFs calculated by the VI Metrics node but I'll be adding that soon.

[Ole Marius Svendsen]

Good Evenig Ryan, i hope you are doing well.

I will be the first to admit with regards to VSC that i am surely have fallen victim for preferred Methode which is VSC. However, in Norway we do use VSC actively as we have experience-based threshold values on towards early facade design and probability for how good access's daylight calculation might be when investigating the facades VSC value. The way we use it is that when applying it on a facade it is easy to see effects of balcony, obstruction elements and how neighbor building impacts the viability of the sky as zenith yields 3x more light compared to horisontal light. 

Of course, i will sympathize with your position as a developer, it is you who most prioritizes your time and development. A new user inquiring more can be challenging as how to respond with respect and firmness of your valuable time. So in this manner i apologize.

For your feedback on csv output. Is the route on towards direct excel export for room name, and general daylight average a potential to see one day? 

take care Ryen, 
Ole m.

[VI-Suite]

Then as far as I can tell, as the DF sky created by the VI-Suite is the recommended CIE overcast sky for VSC you just need to do a DF calculation with a custom radiance parameter of -ab 1 set in the LiVi Simulation node to give you the direct vertical illuminance on your vertical sensing surface divided by the horizontal overcast sky illuminance. Try that against your usual software and let me know how different they are. 

[Ole Marius Svendsen]

Ive done a test now. Same building, same geometry, nothing has been altered.

As you adviced i have change the parameter to -ab 1.

I have made sure i can see the geometry in the preview.

Then i ran the comparison:

And here is the values:

As seen, it is a difference between the two softwares. As VI-Suite allows more tweaking of parameters, i am sure i have done something inaccurately. De Liminæ seem to run his test with the following setting:

-ab 1 -av 0 0 0 -ad 16384 -as 8192 -aa 0.1 -lw 2e-3

[VI-Suite]

Interesting. Can you try with the custom parameters De Luminae uses? The ad and as values especially are a lot higher than Radiance's default ones.

[VI-Suite]

I just noticed that you need to skip the av option as VI-Suite doesn't use it.

[Ole Marius Svendsen]

Unfortunately when copying the string, the simulation tab disappears. This is why i did not display the results with this setting. As i unfortionaly dont know much of any coding languages, i am bound to missunderstand how to get the same custom setting. What would you say i am doing wrong as i cannot run a simulation with higher setting?

 

[VI-Suite]

It's because you still have the av parameter - see my last post. 

[VI-Suite]

By the way, in the VI-Suite master branch you can now add a VI CSV node to the metrics node to write out average DFs to CSV. As this is master branch, you will however need Blender 4.0. 

[Ole Marius Svendsen]

Evening, i unfortionaly did not see that post, though i have committed a change within the setting now. Appolegiz for late reply, i had my test file at work, so redoing it on my laptop took a bit of time to convert the needed changes to be allowed running the file.

Also, i look forward to test out the export functionality with Blender 4.0. I think i would wait a little bit as BlenderBim seems to have some issues with Blender 4.0. I might install both versions to compromise with this for time beeing.

I am now ready to share that the higher setting is reducing the overall difference, though for some more than others. Comen for both test b and c is a grid size of 0.1x0.1. A test for "Felt 6" case "c" with a grid size 0.05x0.05 yielded this time a VSC = 5,19 compared to 5,19.

I am more than happy to run more test if you like if you find this helpful.

[VI-Suite]

I downloaded the trail version of de luminae and confirmed the numbers are the same on simple geometry, so there is a difference in how the VI-Suite and de luminae are treating your geometry.

If you can, preview the de luminae scene with Radiance to see what geometry Radiance sees. 

Looking at your results, windows 5,6,7 all get the same and largest differences, and my feeling is the balcony railing may have something to do with that. 

Comparing windows 2 and 4 in VI-Suite shows an anomaly, as one sees a greater shading effect of the mullion than the other. 

There may also be differences caused by your application of sensing materials in the VI-Suite scene, but without access to the geometry it is difficult to debug.

So, unless further info comes along, I'm considering VSC calculation complete but you may need to do some more investigation to work out the differences in that specific model.

Testing is always helpful and appreciated.

Ryan

[Ole Marius Svendsen]

I will look into the radiance for De luminae as soon as i can and share the results with you.

If you wish, i can send you both the Sketchup file, and the blender file for this test I've committed. I am happy to share and assist, as it is project files from our client i can share it to a email with WeTransfer.

[VI-Suite]

Yes, send them to me.

I'll check them when I have a mo.

[Ole Marius Svendsen]

I tried to upgrade his plugin from 14.0.44 to 14.0.75 as he has had some bug fix for VSC, and use the new radiance engine. This has reduced some of them by 0,07 in value. THough still same main results as initial test.

Also, i found that field 8 and 9 was named inaccurate and thus their results has been swaped.

I will still find time one of these days to look into radiance image.

Blender file format:
https://we.tl/t-JiD83xS61X

Sketchup File Format for DL LIght:
https://we.tl/t-2Yr0wv5uQk

What more can i do to assist?

[Ole Marius Svendsen]

Links as described in the last post. Making radiance image took a lot less time than i thought. Here it is:

[VI-Suite]

Your sketchup model had no surrounding buildings in it, but the Vi-Suite model did. I removed the extra buildings from the VI-Suite model (although they should be there as far as I can tell for a VSC calculation) and the results are the same as below. I also turned off the ground reflectivity in the VI-Suite as you don't want that for VSC. 

1 Felt 1 Average DF 10.82
1 Felt 2 Average DF 12.5
1 Felt 3 Average DF 11.69
1 Felt 4 Average DF 13.29
1 Felt 5 Average DF 7.31
1 Felt 6 Average DF 7.56
1 Felt 7 Average DF 9.51
1 Felt 8 Average DF 14.92
1 Felt 9 Average DF 16.38

[Ole Marius Svendsen]

Evening Ryan. I most say i feel deeply ashamed for something as basic as making sure obstruction buildings is on/off. This is some work Methode i should have been aware off from the start.

with regards to ground to be off, thank u, i was not aware of that beeing on with the settings i had.

All in all, this is really promising results!

[VI-Suite]

Well, we've all done it.

As a suggested methodology when simulation large models, put your sensing surfaces in their own top-level collection. After simulation turn off all other collections and then press Radiance Display. Change what you want in terms of surface visualisation and then turn on the other collections back on. Blender does a full scene update when adding the materials to the sensor surfaces, and this takes a long time with large models.